wave-plan

Decompose complex multi-step tasks into a dependency-aware Directed Acyclic Graph for parallel subagent execution.

2|Updated Jul 22, 2026
One-click install
npx skills add https://github.com/0xUrsanomics/utopia-os --skill wave-plan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wave-plan
Source: https://github.com/0xUrsanomics/utopia-os/tree/main/skills/wave-plan
Command: npx skills add https://github.com/0xUrsanomics/utopia-os --skill wave-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the inefficiency of executing complex, multi-step tasks sequentially, which often leads to bloated context windows and wasted time.

Core Features & Use Cases

  • Dependency-Wave Execution: Automatically decomposes complex asks into a Directed Acyclic Graph (DAG) and groups tasks into parallel waves.
  • Context Management: Keeps the main orchestrator lean by offloading sub-tasks to independent subagents with fresh context.
  • Use Case: When auditing multiple repositories or building a feature that requires parallel research, testing, and documentation, this skill ensures tasks are executed in the most efficient order possible.

Quick Start

Invoke the wave-plan skill by asking the system to plan out a complex multi-step project that involves several independent sub-tasks.

Frequently Asked Questions about wave-plan

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I orchestrate parallel subagents for workflow automation?

Workflow automation with parallel execution is achieved by decomposing complex multi-step tasks into a dependency-aware Directed Acyclic Graph (DAG) and grouping them into topologically sorted waves for concurrent subagent execution.

What is a dependency-aware DAG and when do I need it for task delegation?

A dependency-aware DAG maps task relationships to enable parallel execution. You need it for concurrent research, multi-repo audits, or batch processing of independent sub-tasks to avoid sequential execution bottlenecks and context window bloat.

How do I execute multiple independent sub-tasks in parallel waves?

Executing sub-tasks in parallel waves requires decomposing the overall task into a DAG, grouping independent nodes into execution waves, and offloading them to subagents with fresh context for concurrent processing.

Does parallel task execution help with orchestrator context window optimization?

Parallel task execution optimizes orchestrator context by keeping the main orchestrator lean and offloading sub-tasks to independent subagents that operate with fresh, isolated context windows rather than accumulating state in the main thread.

What is the best way to audit multiple repositories concurrently?

The best way to audit multiple repositories concurrently is using a wave-based orchestration approach that treats each repository audit as an independent sub-task within a DAG, executing them in parallel waves via delegated subagents.

What are the limitations of using a DAG for structured task delegation?

Using a DAG for task delegation limits execution to dependency-resolved parallel waves, meaning tasks with circular dependencies cannot be processed, and all sub-tasks must be clearly decomposable into independent or topologically sortable steps.